Estimating the Phanerozoic history of the Ascomycota ...

https://helda.helsinki.fi

Estimating the Phanerozoic history of the Ascomycota lineages

: Combining fossil and molecular data

Beimforde, Christina

2014

Beimforde , C , Feldberg , K , Nylinder , S , Rikkinen , J , Tuovila , H , Doerfelt , H , Gube , M

, Jackson , D J , Reitner , J , Seyfullah , L J & Schmidt , A R 2014 , ' Estimating the

Phanerozoic history of the Ascomycota lineages : Combining fossil and molecular data ' ,

Molecular Phylogenetics and Evolution , vol. 78 , pp. 386-398 . https://doi.org/10.1016/j.ympev.2014.04.024

http://hdl.handle.net/10138/224253

https://doi.org/10.1016/j.ympev.2014.04.024

cc_by_nc_nd

publishedVersion

Downloaded from Helda, University of Helsinki institutional repository.

This is an electronic reprint of the original article.

This reprint may differ from the original in pagination and typographic detail.

Please cite the original version.

Molecular Phylogenetics and Evolution 78 (2014) 386–398

Contents lists available at ScienceDirect

Molecular Phylogenetics and Evolution

journal homepage: www.elsevier .com/locate /ympev

Estimating the Phanerozoic history of the Ascomycota lineages:Combining fossil and molecular data

http://dx.doi.org/10.1016/j.ympev.2014.04.0241055-7903/� 2014 The Authors. Published by Elsevier Inc.This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

⇑ Corresponding author.E-mail address: [email protected] (C. Beimforde).

Christina Beimforde a,⇑, Kathrin Feldberg b, Stephan Nylinder c, Jouko Rikkinen d, Hanna Tuovila d,Heinrich Dörfelt e, Matthias Gube e,f, Daniel J. Jackson a, Joachim Reitner a, Leyla J. Seyfullah a,Alexander R. Schmidt a

a Courant Research Centre Geobiology, University of Göttingen, Goldschmidtstraße 3, 37077 Göttingen, Germanyb Systematic Botany and Mycology, Faculty of Biology, University of Munich (LMU), Menzinger Str. 67, 80638 Munich, Germanyc Department of Botany, Swedish Museum of Natural History, P.O. Box 50007, SE-104 05 Stockholm, Swedend Department of Biosciences, University of Helsinki, P.O. Box 65, FIN-00014 Helsinki, Finlande Microbial Communication, Friedrich Schiller University Jena, Neugasse 25, 07743 Jena, Germanyf Department of Soil Science of Temperate Ecosystems, Büsgen Institute, University of Göttingen, Büsgenweg 2, 37077 Göttingen, Germany

a r t i c l e i n f o a b s t r a c t

Article history:Received 16 April 2013Revised 12 December 2013Accepted 21 April 2014Available online 30 April 2014

Keywords:AmberAscomycotaBEASTDivergence times estimatesFossil constraintsFungi

The phylum Ascomycota is by far the largest group in the fungal kingdom. Ecologically important mutu-alistic associations such as mycorrhizae and lichens have evolved in this group, which are regarded as keyinnovations that supported the evolution of land plants. Only a few attempts have been made to date theorigin of Ascomycota lineages by using molecular clock methods, which is primarily due to the lack ofsatisfactory fossil calibration data. For this reason we have evaluated all of the oldest available ascomy-cete fossils from amber (Albian to Miocene) and chert (Devonian and Maastrichtian). The fossils representfive major ascomycete classes (Coniocybomycetes, Dothideomycetes, Eurotiomycetes, Laboulbeniomyce-tes, and Lecanoromycetes). We have assembled a multi-gene data set (18SrDNA, 28SrDNA, RPB1 andRPB2) from a total of 145 taxa representing most groups of the Ascomycota and utilized fossil calibrationpoints solely from within the ascomycetes to estimate divergence times of Ascomycota lineages with aBayesian approach. Our results suggest an initial diversification of the Pezizomycotina in the Ordovician,followed by repeated splits of lineages throughout the Phanerozoic, and indicate that this continuousdiversification was unaffected by mass extinctions. We suggest that the ecological diversity within eachlineage ensured that at least some taxa of each group were able to survive global crises and rapidlyrecovered.� 2014 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/3.0/).

1. Introduction

The Fungi constitute a major group of eukaryotic organisms(Hawksworth, 1991, 2001). They exhibit a broad variety of life-styles and morphologies ranging from single celled organisms tomulti-cellular colonies which can be among the largest and possi-bly oldest organisms on earth (Brazee et al., 2012). Most aquaticand terrestrial ecosystems are occupied by a diverse range of fun-gal species. With over 64,000 described species in approximately6400 genera, the phylum Ascomycota is by far the largest phylumin the fungal kingdom (Kirk et al., 2008; Blackwell, 2011). Theautapomorphy of this group is a sack-like structure, the ascus, inwhich the sexual spores are produced. Species of the Ascomycota

are extremely variable in morphology and ecology. As degradersof persistent organic materials such as lignin and keratin, ascomy-cetes play an important role in nutrient cycling. Additionally, manyascomycetes participate in symbiotic associations includingmycorrhizae and lichens.

Phylogenetic relationships among major groups of the Pezizo-mycotina have been the subject of many recent studies (e.g. Liuand Hall, 2004; Lutzoni et al., 2004; Spatafora et al., 2006;Schoch et al., 2009a; Miadlikowska et al., 2006; Hibbett et al.,2007; Ebersberger et al., 2012; Kumar et al., 2012; Morgensternet al., 2012). Several attempts have also been made to date the ori-gin and subsequent evolution of main fungal lineages by molecularclock methods (e. g. Heckman et al., 2001; Sanderson, 2003a;Berbee and Taylor, 1993, 2007; Taylor and Berbee, 2006;Padovan et al., 2005; Lücking et al., 2009, Berbee and Taylor,2010; Gueidan et al., 2011; Floudas et al., 2012; Ohm et al., 2012,

http://crossmark.crossref.org/dialog/?doi=10.1016/j.ympev.2014.04.024&domain=pdf

http://creativecommons.org/licenses/by-nc-nd/3.0/

http://dx.doi.org/10.1016/j.ympev.2014.04.024

http://creativecommons.org/licenses/by-nc-nd/3.0/

mailto:[email protected]


http://www.sciencedirect.com/science/journal/10557903

http://www.elsevier.com/locate/ympev

C. Beimforde et al. / Molecular Phylogenetics and Evolution 78 (2014) 386–398 387

Amo de Paz et al., 2011, Prieto and Wedin, 2013). Fungi probablyderived from aquatic ancestors and diverged at a relatively earlystage during the evolution of the Eukaryota (e.g. Steenkampet al., 2006; Liu et al., 2009; Lara et al., 2010). Their time of diver-gence, however, is still a matter of debate. Simon et al. (1993) werethe first to apply a molecular clock to a fungal phylogeny. Subse-quently, Heckman et al. (2001) estimated that Fungi had occupiedterrestrial habitats for at least 1000 million years, an estimatewhich was revised by Sanderson (2003a). However, these studiesdid not consider substitution rate variation, a phenomenon nowknown to be common in many organism lineages. The existenceof such variation in the fungal phylogeny was demonstrated byBerbee and Taylor (1993, 2010) and is a challenging problem, evenunder the assumption of relaxed clock models, which are able toaccommodate variable substitution rates across individualgroups and genes (e.g. Sanderson, 2003b; Drummond et al.,2006; Drummond and Rambaut, 2007). Considering the numberof ascomycete species and the broad range of morphologies andlife-forms they possess, substitution rate heterogeneity is likelyto be quite drastic across their phylogeny, even at the class level(Lutzoni and Pagel, 1997; Woolfit and Bromham, 2003; Lumbschet al., 2008). Besides improving analytical methods of molecularevolution, the integration of fossil evidence of individual fungal lin-eages would help to partly overcome this problem (Berbee andTaylor, 2010). Many other studies of molecular evolution showedthe importance of constraining molecular clocks with fossil evi-dence (Benton et al., 2009; Hedman, 2010; Inoue et al., 2010;Magallon, 2010; Pyron, 2010; Wilkinson et al., 2011; Lukoscheket al., 2012; Sauquet et al., 2012). A crucial requirement for theuse of fossils as minimum age constrains is their accurateplacement to specific nodes in the phylogeny under study(Rutschmann et al., 2007; Marshall, 2008; Forest, 2009; Parhamet al., 2012; Pyron, 2010; Dornburg et al., 2011). Reliable assign-ment of fossil taxa to modern phylogenies requires accurate infor-mation about their systematic position and age. In this regard,fossilized Fungi preserved in amber and chert are excellent mate-rial as they conserve even delicate microstructures regardless oftheir susceptibility to decay (Stankiewicz et al., 1998; Martínez-Delclo9s et al., 2004). This allows the precise assignment of fossildata to specific phylogenetic nodes.

In order to test the potential use for molecular evolution modelsof Fungi we have evaluated 13 extraordinarily well preserved andprecisely dated fossil ascomycetes, which represent the oldest fos-sil representatives of their respective lineages (see Table 1). Thefossil Fungi are preserved in amber from various deposits spanningan Albian to Miocene age (about 100–17 million years old) as wellas in Devonian and Maastrichtian cherts (about 410 and 66.5 mil-lion years old, respectively).

Here we have assembled a multi-gene data set with a total of145 modern taxa including representatives of most Ascomycotaclasses, and utilized five fossils of Pezizomycotina from amberand chert to estimate divergence times of the main classes. Wehave explicitly used fossil ascomycetes as minimum age con-straints to avoid using secondary node calibrations (age estimatesfrom previous studies). For comparison and to evaluate the influ-ence of our internal node constraints, we also performed an analy-sis with identical parameter settings but with Paleopyrenomycitesas the sole constraint for Pezizomycotina.

This is the first study that evaluates all available fossil ascomy-cetes that represent the oldest reliable evidence of respectiveextant lineages and discusses their suitability as minimum ageconstraints for molecular evolution studies. Five fossils were suit-able for serving multiple calibration points within our dataset.Our results show that the integration of minimum age constraintsin terminal groups of ascomycete classes significantly affects theestimated divergence times of both early branching nodes and

nodes of terminal groups of Ascomycota lineages by pushing themback in time. According to our results the diversification of the Pez-izomycotina started in the Ordovician, followed by a continuousdiversification throughout the Phanerozoic that was likely unaf-fected by mass extinctions.

2. Material and methods

2.1. Fossil ascomycetes from amber and chert

Specimens of all available fossil ascomycetes from amber andchert representing the oldest fossil evidence of their respective lin-eages (Table 1) were reinvestigated considering their potential useas minimum age constraints in molecular models, following theguidelines provided by Parham et al. (2012).

2.1.1. Fossil ConiocybomycetesA well-preserved specimen of Chaenotheca preserved in Baltic

amber (50 – 35 Ma; Fig. 1h; Rikkinen, 2003) clearly belongs tothe Coniocybaceae. Until recently, the phylogenetic position of thisfamily remained enigmatic (Tibell, 2001; Tibell and Koffman,2002). Prieto et al. (2013) proposed that this group of mazaediatefungi is an early diverging group in the inoperculate ascomycetesand defined new the class and order Coniocybomycetes, Coniocyb-ales. Our data strongly support the findings of Prieto et al. (2013).

2.1.2. Fossil DothideomycetesSeveral fossils from Mesozoic and Cenozoic amber deposits clo-

sely resemble extant species of the genus Metacapnodium (Meta-capnodiaceae, Capnodiales; Schmidt et al., 2014). These Fungibelong to the sooty moulds, a term that is commonly used for anecological group of saprophytic Fungi that live on the surfaces ofliving plants. Hyphae of Metacapnodium have a characteristicgrowth form with subglobose cells and gradually tapering apices.The oldest fossil representative of the Metacapnodiaceae isenclosed in Early Cretaceous Charentes amber from France datingabout 100 Ma (Fig. 1b).

Distinctive conidiophores and a plethora of septate, mostlyfour-celled and slightly curved conidia are enclosed in a piece ofEthiopian amber (95–93 Ma; Fig. 1c; Schmidt et al., 2010a). Thestructures are very similar to those of the extant genus Curvularia(Pleosporaceae, Pleosporales) but could also represent a speciesof some other genus in the family (e.g. Drechslera, Bipolaris,Exserohilum). For this reason the authors did not assign the fossilto a modern genus and introduced the new fossil genusPalaeocurvularia.

The fossil parasite Petropus brachyphylli (Fig. 1d) was describedfrom silicified conifer leaves (Brachyphyllum patens; Van der Hamand Van Konijnenburg-van Cittert, 2003) of late Maastrichtianchert (66.5 Ma) by Van der Ham and Dortangs (2005). P. brachy-phylli is considered to be closely related to the extant Phaeocrypto-pus. Phaeocryptopus is very likely polyphyletic and either belongsin the Dothideales or Capnodiales (Zhang et al., 2011; Wintonet al., 2007).

2.1.3. Fossil EurotiomycetesAspergillus collembolorum (Fig. 1f) is preserved in Eocene Baltic

amber (50–35 Ma) and was described by Dörfelt and Schmidt(2005). The fossil includes numerous well preserved conidiophoresvery similar to those of modern species of the Aspergillus flavusgroup (Trichocomaceae, Eurotiales).

Rikkinen and Poinar (2000) described Chaenothecopsis bitterfeld-ensis from Bitterfeld amber (23 Ma; Fig. 1i). Two further specimensof the same genus were described from Eocene Baltic andOligocene Bitterfeld amber dating back to 50–35 Ma and 23 Ma,

Table 1List of ascomycete fossils from amber and chert representing the oldest fossil evidence of their respective lineages, including assignment to extant relatives, repository, referencesfor phylogenetic analyses and age. The fossils are arranged by their age (from old to young).

Fossil taxon Level of assignment Collection Reference forfossildescription

Material andage

Reference for stratigraphy

Paleopyrenomycitesdevonicusa

Pezizomycotina PB 3411, W. Remy collection, Forschungsstelle fürPaläobotanik, Westfälische Wilhelms-Universität,Münster

Taylor et al.(1999, 2005)

DevonianRhynie Chert410 Ma

Richardson (1967), Riceet al. (1995)

Metacapnodiaceaea Metacapnodiaceae,Capnodiales

IGR.ARC-115.3b, Amber collection of GéosciencesRennes at the University Rennes 1

Schmidt et al.(2014)

Charentesamber 100 Ma

Néraudeau et al. (2002),Perrichot et al. (2010)

Palaeocurvulariavariabilis

Pleosporaceae,Pleosporales

MB. Pb.2009/201, Museum für Naturkunde Berlin Schmidt et al.(2010a,b)

Ethiopianamber 95–93 Ma

Schmidt et al. (2010b)

Petropusbrachyphylli

Capnodiales orDothideales

NHMM RD 265, Natuurhistorisch MuseumMaastricht

Van der Hamand Dortangs(2005)

Limburg Chert,Netherlands66.5 Ma

Jagt (1999)

Anzia electraa Anzia, Parmeliaceae,Lecanorales

Oschin 5/0, collection of M. Oschin, Los Angeles Rikkinen andPoinar (2002)

Baltic amber50–35 Ma

Standke (2008)

Aspergilluscollemboloruma

Aspergillus,Trichocomaceae,Eurotiales

No. 805, collection of C & W Hoffeins, Hamburg Dörfelt andSchmidt(2005)


Standke (2008)

Calicium sp.a Calicium, Caliciaceae,Lecanorales

GZG.BST.27296, Geoscientific Collections of theGeorg August University, Göttingen (formerlyArnold collection 1294)

Rikkinen(2003)


Standke (2008)

Chaenotheca sp. Chaenotheca,Coniocybaceae,Coniocybales

GZG.BST.27297, Geoscientific Collections of theGeorg August University, Göttingen (formerlyArnold collection 1285)

Rikkinen(2003)


Standke (2008)

Chaenothecopsis sp. Chaenothecopsis,Mycocaliciaceae,Mycocaliciales

GZG.BST.27286, Geoscientific Collections of theGeorg August University, Göttingen

Tuovila et al.(2013)


Standke (2008)

Gonatobotryumpiceae

Pezizomycotina No. F129/BB/F/CJW, collection of J. Wunderlich,Hirschberg an der Bergstraße

Dörfelt andSchmidt(2007)


Standke (2008)

Stigmatomycessuccini

Stigmatomycets,Laboulbeniaceae,Laboulbeniales

Zoologische Staatssammlung München Rossi et al.(2005)

Bitterfeldamber 23 Ma

Blumenstengel et al. (1999),Dunlop (2010)

Parmelia ambraParmeliaisidiiveteris

Parmeliaceae,Lecanorales

AF9-17E and AF9-17B, amber collection of G.O.Poinar, Oregon State University

Rikkinen andPoinar (2000)

Dominicanamber 17 Ma

Iturralde-Vinent and MacPhee (1996), Iturralde-Vinent, 2001

Phyllopsoradominicanus

Phyllopsora,Ramaliaceae,Lecanorales

Poinar B 1–23, amber collection of G.O. Poinar,Oregon State University

Rikkinen andPoinar (2008)

Dominicanamber 17 Ma

Iturralde-Vinent and MacPhee (1996), Iturralde-Vinent (2001)

a The fossils used as minimum age constraints in this study as indicated in Fig 2.

388 C. Beimforde et al. / Molecular Phylogenetics and Evolution 78 (2014) 386–398

respectively (Tuovila et al., 2013). All three fossils clearly belong tothe order Mycocaliciales, which has usually been placed in theEurotiomycetes (e.g. Schoch et al., 2009a).

2.1.4. Fossil LaboulbeniomycetesA well-preserved specimen of this highly specialized lineage

was found in Bitterfeld amber (23 Ma; Rossi et al., 2005) anddescribed as Stigmatomyces succini (Fig. 1k). The fossil fungus isattached to the thorax of a stalk-eyed fly (Prosphyracephala succini,Diopsidae).

2.1.5. Fossil LecanoromycetesSeveral specimens of Anzia (Rikkinen and Poinar, 2002) are pre-

served in Baltic amber dating back 50–35 Ma (Fig. 1e). Some ofthese fossils are morphologically identical to the extant speciesA. japonica which may be the closest living relative. The genusAnzia belongs to the Parmeliaceae (Lecanorales), the largest familyof lichen forming Fungi and is morphologically very similar to thegenus Pannoparmelia (Thell et al., 2012).

Poinar et al. (2000) described two species of Parmelia (P. ambraand P. isidiiveteris) from Dominican amber (17 Ma; Fig. 1l). Bothfossils clearly belong to the family Parmeliaceae, but they cannotbe assigned with certainty to any extant genus. However, neitherof the two fossil species represents Parmelia sensu stricto.

A fossil specimen of the genus Phyllopsora (Ramalinaceae,Lecanorales) preserved in Dominican amber (17 Ma; Fig. 1m) was

described as P. dominicanus by Rikkinen and Poinar (2008). Themorphological features of P. dominicanus closely resemble thosefound in modern Phyllopsora species and are very similar to therecent P. chlorophaea for example.

A fossil representative of the genus Calicium (Rikkinen, 2003) ispreserved in amber of the Baltic deposit dating back 50–35 Ma(Fig. 1g). Species of Calicium (Caliciaceae, Teloschistales) are typical‘‘calicioid lichens’’, a paraphyletic assemblage of Fungi sharingmorphological similarities such as stalked fruiting bodies and apowdery spore mass called the mazaedium (Tibell, 1984).

2.1.6. Fossil ascomycetes of groups with ambiguous systematicpositions

Gonatobotryum piceae (Dörfelt and Schmidt, 2007) is enclosed inBaltic amber (50–35 Ma; Fig. 1j). The fossil specimen shows closesimilarities to modern Gonatobotryum fuscum, but developed dif-ferent conidiophores and mature conidia. Teleomorphs are cur-rently unknown for Gonatobotryum species (Arx, 1981).

Paleopyrenomycites devonicus (Fig. 1a) is by far the oldest evi-dence for ascomycetes. It is enclosed in Devonian Rhynie Chert dat-ing back 410 million years (Taylor et al., 2005). P. devonicus wasoften assigned to Sordariomycetes, but its exact systematic posi-tion is disputed (Taylor et al., 2005; Eriksson, 2005; Padovanet al., 2005; Taylor and Berbee, 2006). An assignment to the Pezizo-mycetes seems also possible, since Paleopyrenomycites might haveproduced operculate asci (Lücking et al., 2009).

Fig. 1. Fossil ascomycetes from amber and chert representing the oldest known ancestors of respective lineages. (a) Perithecia of Paleopyrenomycites devonicus from EarlyDevonian (Pragian) Rhynie Chert. W. Remy collection PB 3411. Courtesy of Hans Kerp (University of Münster). (b) Moniliform hyphae of a Metacapnodiaceae representativefrom Early Cretaceous (Albian) Charentes amber. IGR.ARC-115.3b. (c) Conidia of Palaeocurvularia variabilis from Late Cretaceous (Cenomanian) Ethiopian amber. MB. Pb. 2009/200. (d) Hypostroma of Petropus brachyphylli from Maastrichtian chert from the Netherlands. NHMM RD 265. Courtesy of Raymond W. J. M. van der Ham (NaturalisBiodiversity Center, Leiden). (e) Anzia electra from Eocene Baltic amber. Hoffeins 950-1. (f) Sporulating conidiophore of Aspergillus collembolorum on a springtail from EoceneBaltic amber. Hoffeins 805. (g) Ascoma of Calicium sp. from Eocene Baltic amber. GZG.BST.27296. (h) Ascoma of Chaenotheca sp. on remnant bark in Eocene Baltic amber.GZG.BST.27297. (i) Ascoma of a resinicolous Chaenothecopsis sp. from Eocene Baltic amber. GZG.BST.27286. (j) Sporulating conidiophore of Gonatobotryum piceae on a coniferseedling from Eocene Baltic amber. Wunderlich F129. (k) Three thalli of Stigmatomyces succini on a dipteran from Oligocene Bitterfeld amber. Zoologische StaatssammlungMünchen, sine numero. (l) Parmelia ambra from Miocene Dominican amber. Poinar AF9-17E. Courtesy of George O. Poinar, Jr. (Corvallis). (m) Phyllopsora dominicanus fromMiocene Dominican amber. Poinar B 1-23. Scale bars: 10 lm (b–d, f, and j), 100 lm (a, g–i, k, and m), and 1 mm (e and l).



2.2. Taxon sampling for phylogenetic reconstruction and molecularwork

For this study we used the small and large ribosomal subunit(nucSSU and nucLSU respectively) and RNA polymerase II proteincoding genes RPB1 and RPB2 as implemented in a previous studyby James et al. (2006). Sequences were obtained from culturedstrains ordered from the CBS (Centraalbureau voor Schimmelcul-tures, Utrecht), JMRC (Jena Microbial Resource Collection), andfrom Genbank. Additional Fungi were collected from localities inFinland (2009) and New Caledonia (2011). The resulting taxonset consists of 145 species representing most classes of theAscomycota. Accession numbers of all sequences are provided inSupplementary Table 1. For protein coding and ribosomal genes,we isolated DNA from fungal material using the Invisorb Spin PlantMini Kit (Invitek, Berlin, Germany) and NucleoSpin�Plant DNAextraction kit (Macherey–Nagel) with the following modificationto the manufacturer’s protocol: some specimens were incubatedup to 2 h to ensure the lysis of the ascocarps. PCR reactions werecarried out with fungal specific primers: SSU ribosomal genes wereamplified with the primers NS1, NS2, NS3, NS4 (White et al., 1990)and NS24 (Gargas and Taylor, 1992); LSU ribosomal genes wereamplified with LR0 (Rehner and Samuels, 1994), LR3R (Moncalvoet al., 2000), LR5 and LR7 (Vilgalys and Hester, 1990). Genes codingfor the RNA polymerase II were amplified with the primers RPB1-AFasc, RPB1-6R2asc, RPB1-DF2asc, RPB1-GlRasc and RPB1G2R(Hofstetter et al., 2007) for the largest subunit and fRPB2-5f,FRPB2-7cf, fRPB2-7cR, fRPB2-11aR and RPB2-11bR (Liu et al.,1999) for the second largest subunit. PCR reactions were per-formed according to the protocols listed in respective referencefor mentioned primers. In case of melanin inhibiting the PCR, theDNA-templates were diluted up to 5000 fold sometimes with theaddition of 200 ng/ll bovine serum albumin (BSA) (Kreader,1996). PCR products were purified using PCRapace (Invitek, Berlin,Germany). All PCR products were sequenced in both directionswith a MegaBACE 1000 automated sequencing machine andDYEnamic ET Primer DNA Sequencing Reagent (AmershamBiosciences, Little Chalfont, UK). All sequences were assembledand edited using Bioedit 5.0.9 (Hall, 1999) and Seaview 4 (Gouyet al., 2010).

2.3. Initial phylogenetic Analysis

Datasets for each gene (SSU, LSU, RPB1 and RPB2) were alignedseparately using MAFFT version 6 (Katoh and Toh, 2008) with sub-sequent manual adjustment to minimize the number of possiblefalse homologies using Bioedit 5.0.9. (Hall, 1999) and Seaview 4(Gouy et al., 2010). Unalignable regions and introns were excludedby using the mask function in Bioedit 5.0.9. Best fitting substitutionmodel for each gene was chosen separately from seven substitu-tion schemes included in the software package jModeltest 2.1.1(Darriba et al., 2012), and models were chosen according to theBayesian information criterion (BIC, Schwarz, 1978). The Bayesianinformation criterion supported the TrN + G model as the best fitfor LSU, SYM + G for SSU, and GTR + G for RPB1 and RPB2. Topolog-ical congruence of the four datasets was assessed by visualcomparison of phylogenetic trees obtained from maximumlikelihood-based analysis with RaxML (Stamatakis et al., 2008),and all genes were subsequently combined in a super matrix usingBioedit 5.0.9. Bayesian analyses were carried out using Markovchain Monte Carlo (MCMC) in MrBayes 3.1.2 (Ronquist andHuelsenbeck, 2003) to generate a reasonable starting tree for sub-sequent analyses of divergence date estimates in BEAST. Analyseswere run using four chains for 10 million generations each,sampling parameters every 1000th generation. All analyses wereperformed on the freely available computational resource CIPRES

(www.cipres.org). Average standard deviations of split frequency(ASDSF) lower than 0.01 were interpreted as indicative of indepen-dent MCMC convergence.

2.4. Fossil calibrations

The placement of the fossil Paleopyrenomycites (Fig. 1a) is chal-lenging since its exact systematic position is not clear (Taylor et al.,2005; Lücking et al., 2009; Taylor and Berbee, 2006). The previ-ously discussed possibilities for its placement include anywherein the Pezizomycotina stem lineage, Pezizomycotina crown group,or members of the Pezizomycotina building operculate asci(Lücking et al., 2009). For our evolutionary model we adopted aconservative view and placed Paleopyrenomycites on the crowngroup of Pezizomycetes, thus assuming the common ancestor ofall filamentous, sporocarp-producing Ascomycota (Pezizomycoti-na) to be at least 400 Ma. We decided to model the uncertaintyof the group by applying a truncated normal distribution with anupper hard bound (truncation) set to 400 Ma, corresponding tothe mean of the normal distribution with a standard deviation(SD) of 150, providing an upper 97.5% credibility interval (CI) of700 Ma. This mode of calibration associates an increased uncer-tainty with the immediate upper bound, allowing a more generousinterpretation of the age of a group compared to that of an expo-nential decay. The fossil Anzia electra (Fig. 1e; Rikkinen andPoinar, 2002) was used to calibrate the split between Anzia andother groups of parmeloid lichens here presented by Canoparmeliaconstraining the node to 35 Ma with a truncated normal distribu-tion to model the uncertainty (mean = 35, SD = 50, CI = 135). Basedon the fossil Calicium (Fig. 1g; Rikkinen et al., 2003) we constrainedthe common ancestor of Calicium viride and C. salicium, which areboth morphologically indistinguishable from the fossil to 35 Ma(truncated normal distribution, mean = 35, SD = 50, CI = 135).Using the fossil Aspergillus collembolorum (Fig. 1f; Dörfelt andSchmidt, 2005) we constrained the common ancestor of Aspergillusto 35 Ma (truncated normal distribution, mean = 35, SD = 50,CI = 135). The fossil Metacapnodiaceae (Fig 1b; Schmidt et al.,2014) gave rise to the hypothesis of the common ancestor of theorder Capnodiales to be constrained to an age of 100 Ma (truncatednormal distribution, mean = 100, SD = 150, CI = 400). All analysesof divergence time estimates using the above set of constraintswere first run on empty alignments to check for cross prior influ-ence, while constraining all calibrated nodes to monophyly.

2.5. Divergence time estimates

Subsequent divergence time analyses were carried out usingBEAST 1.7.4 (Drummond et al., 2012). Separate partitions for eachincluded gene were created with BEAUti 1.7.4 (BEAST package). Toaccommodate for rate heterogeneity across the branches of thetree (e. g. Berbee and Taylor, 2010) we used an uncorrelatedrelaxed clock model (Drummond et al., 2006) with a lognormal dis-tribution of rates for each gene estimated during the analyses. Abirth/death tree prior accommodating for incomplete sampling(Stadler, 2009) was used to model the speciation of nodes in thetopology, with uniform priors on probability of splits and extinc-tions. To avoid using uninformative priors on the clock modelswe used vague priors on the substitution rates for each gene (expo-nential decays with mean 0.1 in units of substitutions per site pertime unit). To ensure congruence we ran the analyses five times for100 million generations each, sampling parameters every 25,000generations, assessing convergence and sufficient chain mixing(Effective sample sizes > 200) using Tracer 1.5 (Rambaut andDrummond, 2009). After removal of a proportion of each run asburn-in the remaining trees were combined using LogCombiner(part of the BEAST-package), and summarized as maximum clade


credibility (MCC) trees in TreeAnnotator (part of the BEAST-pack-age), and visualized using FigTree (Rambaut, 2006–9, http://tree.-bio.ed.ac.uk/software/figtree/).

In order to evaluate the possible presence of constant diversifi-cation rate in the evolutionary history of the Ascomycotes, aLineage-through-time (LTT) plot was constructed in Tracer 1.5(Rambaut and Drummond, 2009) from the combined posterior dis-tribution of sampled tree topologies (Fig 3). We also applied theMonte Carlo constant rates (MCCR) test (Pybus and Harvey,2000) to further elucidate this hypothesis.

3. Results

3.1. Topology of the Ascomycota phylogeny

With some exceptions the topologies resulting from the BEASTanalyses (Figs. 2 and S1) are generally congruent with the resultsreported by James et al. (2006) and other large scale phylogeniesof Ascomycota e.g. Schoch et al. (2009a,b). The placement of Pezi-zomycetes basal to Orbiliomycetes is consistent with e.g. Schochet al. (2009a,b) but is opposite in James et al., 2006 and somerecent papers (Ebersberger et al., 2012). Kumar et al. (2012) exten-sively discussed the phylogenetic placement of Orbiliomyceteswith additional ultrastructural analyses that supported the basalstate of Orbiliomycetes in the Pezizomycotina. However, molecularsupport for the node resolving the relationships between the twobasal Pezizomycotina classes (Pezizomyctes and Orbiliomycetes)is low and topologies are unstable.

Geoglossomycetes build the sister group to the Lichinomycetes-Coniocybomycetes clade in our analyses. The genus Leotia, believedto be the most basal member of the Leotiomycetes, here groupstogether with the sister clade of Sordariomycetes with unanimoussupport (1.0 pp). Geoglossomycetes were shown to be a ratherbasal clade close to Pezizomycetes and Orbiliomycetes, and dis-tinct from Leotiomycetes after transfer of a few genera previouslyotherwise assigned (Spatafora et al., 2006, Schoch et al., 2009a,bHustad et al., 2013). Leotiomycetes exclusive of most geoglossoidspecies are currently seen as sister taxon to both Sordariomycetesand Laboulbeniomycetes (Schoch et al., 2009a,b). The placement ofLeotia lubrica was problematic also in previous analyses (Jameset al., 2006) and could possibly be resolved with a higher taxonsampling of Leotiomycetes, which would however exceed thescope of this work.

In our resulting single- and multiple-constrained trees Pezizo-mycetes constitutes the sister clade to remaining representativesof Pezizomycotina with strong support in both analyses (1.0 pp).The Orbiliomycetes build a sister group to Leotiomyceta, but in aposition only indicated by low node support (0.88 vs. 0.91 pp).The Arthoniomycetes–Dothideomycetes-clade has unanimoussupport in both, single- and multi-constrained analyses (1.0 pp).The remaining representatives of the Pezizomycotina split in twoclades. The first clade (0.68 vs. 1.0) consists of the Coniocybomyce-tes together with Lichinomycetes (1.0 vs. 0.99 pp), with the Geo-glossomycetes as sister group (0.99 vs. 0.98 pp). The sister groupto the Geoglossomycetes–Coniocybomycetes–Lichinomycetesclade is the Lecanoromycetes-Eurotiomycetes clade, in both analy-ses with unanimous support (1.0 pp). The second clade consists ofthe Leotiomycetes and Sordariomycetes grouping together withunanimous support in both analyses (1.0 pp).

3.2. Divergence time estimations using five internal calibrations

Divergence time estimates using all five fossil calibrationpoints are also shown in Fig. 2, with horizontal bars representingthe 95% highest posterior density (HPD) intervals for each node.

Comparable results from both analyses (Figs. 2 and S1) are listedin Table 2. According to our data, the Ascomycota diverged fromBasidiomycota in the Neoproterozoic, about 642 Ma (504–859 Ma, 95% HPD interval). The subphylum Pezizomycotina,containing all sporocarp forming members of the Ascomycota(except for Neolecta which resides in Taphrinomycotina and wasexcluded in our analysis as this group or taxon is still poorlyunderstood), split from Saccharomycotina in the early Cambrian,around 537 Ma (443–695). The earliest split in the Pezizomycotina(Pezizomycetes from the remaining Pezizomycotina) occurredin the Ordovician, around 458 Ma (400–583). Within thePezizomycotina, the Orbiliomycetes diverged in the Silurian,430 Ma (353–554). Dothideomycetes + Arthoniomycetes divergedin the Late Devonian, 362 Ma (286–476) and this clade divergedfrom other Letiomyceta in the Early Devonian, around 397 Ma(330–521). Lichinomycetes split from Coniocybomycetes in the inthe Permian, 274 Ma (197–379). Eurotiomycetes and Lecanoromy-cetes divergend in the early Carboniferous, 353 Ma (289–459). Theearliest split in the Eurotiomycetes (Eurotiomycetes crown group)occurred around 336 Ma (273–437) and in the Lecanoromycetes315 Ma (255–414).

4. Discussion

4.1. Systematic assignment of fossil Fungi

A crucial issue in molecular dating studies is the interpretationof morphological characters used to assign fossils to particularnodes in the phylogenies (Rutschmann et al., 2007; Marshall,2008; Forest, 2009; Parham et al., 2012; Pyron, 2010; Dornburget al., 2011; Feldberg et al., 2013). The use of morphological datato reconstruct the evolution of lineages through time can be lim-ited due to homoplasy. Ascomycetes show many cases of parallelevolution in both vegetative and reproductive structures (e.g.Lumbsch, 2000; Schoch et al., 2009a). In this study we have onlyused fossils which we believe to represent extant families or gen-era (with the exception of Paleopyrenomycites which was assignedto Pezizomycotina). However, some level of uncertainty willalways remain when working with fossil material. Besides Paleopy-renomycites devonicus we finalized our selection using fossilsassigned to four extant taxa of ascomycetes (Metacapnodiaceae,Anzia electra, Aspergillus collembolum and Calicium; Table 1, Fig. 1)which provided minimum ages for the split of the lineage fromits sister group. The remaining eight fossils (Fig. 1; Table 1) didnot provide suitable minimum age constraints. This was mainlydue to insufficient taxon sampling of our molecular data. Whilewe could not use all of the 13 available fossils in our study, all ofthem are potentially of value for further studies in Fungi with den-ser taxon samplings, or focus on the evolution of individual groupsof ascomycetes.

Gueidan et al. (2011) used Palaeocurvularia (Fig. 1c; Schmidtet al., 2010a,b) to constrain the split between Arthoniomycetesand Dothideomycetes. However, the fossil is represented bynumerous conidia and related conidiophores, which resemblethose produced by modern species of Curvularia but also those ofBipolaris, Drechslera, and Exserohilum. Since fragments of the possi-ble teleomorph are poorly preserved, and the fossil conidia aremore variable than those of any of the modern genera, Schmidtet al. (2010b) and Gueidan et al. (2011) avoided an assignment ofthe fossil to any modern family. In order to avoid a false assign-ment, we had to exclude Palaeocurvularia from our analyses.

Gonatobotryum piceae (Fig. 1j; Dörfelt and Schmidt, 2007) wasalso excluded, because of the ambiguous phylogenetic positionof this genus (Arx, 1981). The morphologically similar fossil Gona-tobotrys primigenia (Caspary and Klebs, 1907) likely represents a

http://tree.bio.ed.ac.uk/software/figtree/


Fig. 2. Maximum clade credibility (MCC) tree with divergence times estimates for main groups of the Ascomycota obtained from a Bayesian approach (BEAST) using five fossilminimum age constraints. Numbers at nodes indicate posterior probabilities (pp) for node support. Bars correspond to the 95% highest posterior density (HPD) intervals. Forestimated median ages of numbered nodes, see Table 2. Only node supports below 1.0 pp are shown. Assignments in the tree of the fossil minimum age constraints aremarked with red circles. Geological periods are abbreviated as: Cam. = Cambrian, Ord. = Ordovician, Sil. = Silurian, Dev. = Devonian, Carb. = Carboniferous, Perm. = Permian,Tri = Triassic, Jur. = Jurassic.


0100200300400500Time (Ma)

0

10

100

1000

Line

ages

(N)

Fig. 3. Lineage-through-time (LTT) plot constructed from combined posteriordistribution of sampled tree topologies by using Tracer 1.5 (Rambaut andDrummond, 2009).


species of Gonatobotryum rather than Gonatobotrys. As the moderngenus Melanospora (Ceratostomataceae, Sordariomycetes) isknown as the teleomorph of Gonatobotrys (Vakili, 1989), this genuscould potentially be used for calibration. A confident assignment ofGonatobotrys primigenia would require a re-investigation, however,this fossil which was part of the Künow collection of Berlin’sMuseum of Natural History is lost without any trace. Both fossilsare well preserved and might serve as calibration constraints oncetheir position within the ascomycetes has been clarified.

Petropus brachyphylli (Fig. 1d; Van der Ham and Dortangs, 2005)was not used in our calculation because of the uncertain taxonomicplacement of the corresponding modern genus Phaeocryptopus,which is likely polyphyletic and either belongs to the Dothidealesor Capnodiales (Zhang et al., 2011; Winton et al., 2007).

Stigmatomyces succini (Fig. 1k; Rossi et al., 2005) was also notused in our study although the fossil is well dated and confidentlyassigned to the genus Stigmatomyces (Laboulbeniomycetes). Spe-cies of this ectoparasite class display distinct morphologies andtheir phylogenetic position has long been unclear, but Schochet al. (2009a) have recently proposed a sister relationship of Labo-ulbeniomycetes to Sordariomycetes. Primary analyses includingsequences of the Laboulbeniomycetes indicated the introductionof substantially long branches in resulting phylogenies (data notshown), and this class was therefore excluded from further

Table 2Divergence time estimates of Ascomycota lineages obtained from Bayesian analysis using eifrom amber (Metacapnodiaceae, Anzia electra, Aspergillus collembolorum, and Calicium; Tableprovided. Divergence times are provided in millions of years (Ma). The node numbers cor

Nodes One calibration

Geological period

1 Ascomycota crown group –2 Pezizomycotina–Saccharomycotina –3 Saccharomycotina crown group –4 Pezizomycotina crown group Ordovician5 Pezizomycetes crown group Devonian6 Orbiliomycetes–other Pezizomycotina Devonian7 Arthoniomycetes–Dothideomycetes Carboniferous8 Dothideomycetes crown group Carboniferous9 Leotiomycetes–Sordariomycetes Permian10 Sordariomycetes crown group Triassic11 Lichinomycetes–Coniocybomycetes Triassic12 Eurotiomycetes crown group Permian13 Lecanoromycetes crown group Permian14 Eurotiomycetes–Lecanoromycetes Carboniferous

analyses to avoid introducing unnecessary bias into the branchlength estimates.

The fossils Parmelia ambra (Fig. 1l) and P. isidiiveteris (Poinaret al., 2000) cannot be assigned with confidence to particular gen-era within the foliose parmelioid lichens (‘‘Parmelia sensu lato’’).We were unable to use Parmelia because it would imply a con-straint on the divergence of Parmelia and Canoparmelia, or Anzia,to a minimum of 17 Ma. As we had already used Anzia electra(50–35 Ma; Rikkinen and Poinar, 2002) to constrain the split ofAnzia and Canoparmelia, an integration of the much younger fossilof parmelioid lichens would introduce redundancy.

Similar reasons led to the exclusion of Phyllopsora dominicanus(Fig. 1m; Rikkinen and Poinar, 2008) as an age constraint in ouranalysis. Our Phyllopsora sequences grouped together with Bacidiaand constraining the divergence between these two genera to aminimum of only 17 Ma would not have been realistic (Printzenand Lumbsch, 2000; Rikkinen and Poinar, 2008).

The fossils Chaenotheca (Fig. 1h; Rikkinen, 2003), andChaenothecopsis (Fig. 1i; Rikkinen and Poinar, 2000; Tuovila et al.,2013) were discarded from the analyses despite being of excellentquality. Initial tests for cross-prior influence on the age estimatesof nodes indicated that the introduction of these constraintsresulted in several other constraints showing bimodal posteriordistributions. Finding the underlying cause of such phenomenacan be difficult, but is possibly due to discordance between thefit of the tree prior and one or more node constraints. Removal ofthe mentioned fossils indicated substantial performance improve-ments across the tree, validating the decision for removal.

4.2. The impact of internal node constraints on estimated divergencetimes of Pezizomycotina classes

Some studies have evaluated the variation resulting from differ-ent calibration strategies in fungal phylogenies (Taylor and Berbee,2006; Lücking et al., 2009; Padovan et al., 2005), but none of themevaluated the impact of internal node constraints on models offungal molecular evolution. Compared to the sole use of the Devo-nian Paleopyrenomycites, the use of four calibrations from Mesozoicand Cenozoic Pezizomycotina crown group fossils (of Dothideomy-cetes, Eurotiomycetes and Lecanoromycetes) in addition to theDevonian fossil resulted in older age estimates (Table 2). Usingmultiple age constraints only slightly affected the first split inthe Pezizomycotina from 444 Ma (400–576, when using onlyPaleopyrenomycites) to 458 Ma (400–583) when using the fouradditional calibrations. All other Pezizomycotina have diverged

ther Paleopyrenomycites as single calibration or together with 4 additional calibrations1, Fig. 1). For each divergence, the median and the 95% Highest Posterior Density are

respond to numbers used in Fig. 2 to show their placement in the chronogram.

Five calibrations

Time (Ma) Geological period Time (Ma)

– Neoproterozoic 588 (487–773)– Cambrian 537 (443–695)– Devonian 373 (276–514)444 (400–576) Ordovician 458 (400–583)408 (262–543) Devonian 413 (292–554)407 (328–534) Silurian 430 (353–554)335 (263–450) Devonian 362 (286–476)321 (247–427) Carboniferous 350 (273–459)287 (234–388) Carboniferous 309 (267–430)233 (182–316) Permian 260 (207–339)246 (164–355) Permian 274 (197–379)282 (215�382) Carboniferous 336 (273–437)260 (190–356) Carboniferous 315 (255–414)327 (250�436) Carboniferous 353 (289–459)


from the Orbiliomycetes in the Silurian 430 Ma (353–554), ratherthan in the Early Devonian some 407 Ma (328–534). Both analysesresulted in congruent relationships between Eurotiomycetes, Lec-anoromycetes, Dothideomycetes and Lichinomycetes allowing acomparison of the divergence times of these Pezizomycotina clas-ses; these divergence estimates are significantly older when usingadditional age constraints (Table 2).

Besides affecting the divergence times of early Ascomycota lin-eages, the integration of additional age constraints resulted inolder age estimates of more recent ascomycete groups (terminalnodes). These effects are not only restricted to branches associatedwith fossil age constraints, although adjacent branches are slightlystronger affected, supporting the observations of Berbee and Taylor(2010).

Our results show that the use of fossil age constraints (even ifrelatively young) in terminal groups of ascomycetes significantlyaffects the estimated divergence times of both early branchingnodes and terminal groups of Ascomycota lineages. This effectwas also observed when using different BEAST parameters, e.g.unconstrained uniform probability distributions to model ageuncertainties of groups associated with fossils (data not shown).

4.3. Comparisons to previous studies

Compared to earlier studies our data indicate either muchyounger (Heckman et al., 2001) or much older (Berbee andTaylor, 1993) age estimates of Ascomycota lineages. Our resultsare generally more congruent with the estimates of recent studies(Padovan et al., 2005; Taylor and Berbee, 2006; Lücking et al.,2009; Gueidan et al., 2011; Prieto and Wedin, 2013) (Table 3).One likely explanation is that molecular clock methods haveimproved by developing relaxed molecular clock models, whichallow for more flexible modeling of rate heterogeneity across phy-logenetic trees (e. g. Sanderson, 2003b; Drummond et al., 2006,2012). Additionally, more well resolved fungal phylogenies haverecently been established (e. g. Spatafora et al., 2006; Schochet al., 2009a; Miadlikowska et al., 2006; Hibbett et al., 2007;Ebersberger et al., 2012; Kumar et al., 2012; Morgenstern et al.,2012). Advances in both fields of research have enabled the estab-lishment of increasingly realistic models of evolution for Fungicompared to earlier studies, resulting in different age estimates(e.g. Simon et al., 1993; Heckman et al., 2001).

Table 3Comparison of divergence time estimates from our analyses with previous studies (HeckmGueidan et al., 2011; Prieto and Wedin, 2013) in millions of years (Ma) which also used Palisted. Node numbers correspond to numbers used in Fig. 2 and Fig. S1 to show their placem(w) indicates studies that also used external (non-ascomycotan or non-fungal) calibration

Nodes This study Prieto & Wedin(2013) (Scen. 4)

Gue(201

Cali. 1(1 fossil)

Cali. 2(5 fossils)

1 Ascomycota crown – 588 531 5382 Pezizomycotina–Saccharomycotina – 537 – –3 Saccharomycotina crown – 373 – –4 Pezizomycotina crown 444D 458D 485D 4555 Pezizomycetes crown 408 413 322 �316 Orbiliomycetes–other Pezizomycotina 407 430 – –7 Dothideomycetes–Arthoniomycetes 335 362 302 3628 Dothideomycetes crown 321 350 174 3389 Leotiomycetes–Sordariomycetes 287 309 247 �3410 Sordariomycetes crown 233 260 130 29911 Lichinomycetes-Coniocybomycetes 246 274 267 –12 Eurotiomycetes crown 282 336 306 34113 Lecanoromycetes crown 260 315 305 32214 Eurotiomycetes–Lecanoromycetes 327 353 �320 371

Despite these improvements in methodology and data sam-pling, age estimates are not fully consistent across recent studies.Such discrepancies are likely to have various reasons such asinability to properly model evolutionary rates, parameter settingsfor the applied relaxed clock models, unequal taxon sampling,and choice of genes under study. Such differences make it difficultto compare inferred age estimates of individual studies. Anotherpossible source for inconsistent age estimates in earlier studies isthe assignment of the fossil Paleopyrenomycites devonicus (Tayloret al., 1999, 2005). This fossil constitutes a highly influential con-straint, and since it became available has been used in all studiesof fungal molecular evolution (Table 3). In early studies the fossilwas interpreted as belonging to the Sordariomycetes (e.g.,Heckman et al., 2001). More recent studies used Paleopyrenomy-cites to calibrate the Pezizomycetes crown or stem group (Taylorand Berbee, 2006; Lücking et al., 2009), the Ascomycota crowngroup (Taylor and Berbee, 2006), or as a constraint for the splitbetween Leotiomyceta and other Pezizomycotina (Gueidan et al.,2011) due to the putative operculate ascus. However, the apparentoperculate opening might also be a diagenetic phenomenon(Lücking et al., 2009). Lücking et al. (2009) provide a comprehen-sive discussion concerning the placement of this fossil while recal-ibrating several earlier studies (Berbee and Taylor, 1993; Simonet al., 1993; Doolittle et al., 1996; Redecker et al., 2000;Heckman et al., 2001; Padovan et al., 2005) by reassessing the sys-tematic placement of Paleopyrenomycites. However, Taylor andBerbee (2006) convincingly showed the placement of this fossilat different positions in the Ascomycota tree (Ascomycota crowngroup, Pezizomycotina crown group, Sordariomycetes crowngroup) to have a dramatic effect on estimated ages of fungal lin-eages. Therefore our age estimates are best comparable to otherstudies using Paleopyrenomycites as constraint for the Pezizomyco-tina crown group (e.g., Taylor and Berbee, 2006; Lücking et al.,2009). Our resulting age estimates from the calibrations usingPaleopyrenomycites as a sole constraint are overall consistent withthe ages inferred by Lücking et al. (2009; Table 3) and those ofTaylor and Berbee, 2006; calib. 2 and calib. 3 in Table 3).

Divergence times estimates obtained from our analysis thatemployed five internal calibration points correspond most closelyto those of Gueidan et al. (2011), Prieto and Wedin (2013) andLücking et al. (2009). Gueidan et al. (2011) used Paleopyrenomycitesto calibrate the Pezizomycetes-Leotiomyceta split (whichcorresponds to the Pezizomycotina crown group with the excep-

an et al., 2001; Padovan et al., 2005; Taylor and Berbee, 2006; Lücking et al., 2009;leopyrenomycites as fossil age constraints. For divergence times, only the medians are

ent in the chronogram. Triangle (D) marks the assignment of Paleopyrenomycites; psis; phi (U) recalibration study.

idan et al.1)w

Lücking et al.(2009)U

Taylor and Berbee(2006)

Padovanet al. (2005)

Heckman et al.(2001)w

Cali.1

Cali.2

Cali.3w

Cali.4

Cali.1w

Cali.2w

– 1316 745 652 400D 1148 724 1144400–520D – – – – 1072 657 1085– – – �140 �90 – – 841320–400D 707 400D 400D 215 972 569 –

0 – – – – – �900 �500 –– – – – – – – –– – – – – – – –– – – – – – – –

0 290–280 – – – – – – –400D 226 211 122 653D 400D 400D

– – – – – –270–350 – – – – – – –280–330 – – – – 816 453 –– – – – – – – –


tion of Orbiliomycetes) and Anzia electra for calibrating the splitbetween Anzia and Canoparmelia. Additionally, they also utilizeda metacapnodiaceous fossil and Palaeocurvularia for the split ofDothideomycetes and Arthoniomycetes, together with severalnon-ascomycotan (Taylor et al., 1994; Redecker et al., 2000;Hibbett et al., 1995, 1997), and non-fungal (Crane et al., 1995;Douzery et al., 2004) calibration constraints. Compared to theirstudies, our data include more ascomycotan calibration pointsand no external constraints, which resulted in older ages for someof the ascomycete lineages.

Prieto and Wedin (2013) also utilized fossil age constrainssolely from within the Ascomycota. With the exception of Paleopy-renomycites (Taylor et al., 1999, 2005) and the fossil Calicium(Rikkinen, 2003) they utilized a different set of additional age con-straints. Prieto and Wedin (2013) relied on the fossils Alectoria suc-cina (Mägdefrau, 1957) and Parmelia ambra (Rikkinen and Poinar,2000), which were excluded in our study because of insufficientinformation for assigning these fossils to a corresponding moderngenus (compare Section 4.1). The fossil Alectoria is only poorly pic-tured in a short paper by Mägdefrau (1957) and is not available forre-evaluation. It was part of the private collection of A. Scheele(Allgäu, Germany) but has seemingly been lost forever. Westrongly object to including doubtful fossils, which could introduceerroneous information and result in highly biased age estimates. Incontrast we utilized the fossil Anzia electra (Rikkinen and Poinar,2002), which reveals much better preservation. In contrast toPrieto and Wedin (2013) we avoided integrating the fossils Chae-notheca (Rikkinen, 2003) and Chaeonothecopsis (Tuovila et al.,2013) as they caused cross-prior influence strongly affecting othernode age estimates (compare Section 4.1). Additionally, Prieto andWedin (2013) did not include the influential metacapnodialeanfossils described by Rikkinen et al. (2003) and Schmidt et al.(2014) and the extraordinarily well preserved Aspergillus collembo-lorum (Dörfelt and Schmidt, 2005).

However, the general congruence of recent studies using com-parable parameter settings indicates an increase in convergenceof age estimates. Our results indicate that further inclusions of reli-able fossil constraints are likely to lead to even more accurate esti-mated ages of individual lineages.

4.4. Reconstruction of the evolutionary history of ascomycete lineages

According to our results (Fig. 2), all Pezizomycotina classesoriginated in the Phanerozoic, while the main diversificationbegan in the Ordovician with the divergence of Pezizomycetes(the earliest branching class of Pezizomycotina) from the remain-ing Pezizomycotina. Based on the results of our study it is impos-sible to draw any certain conclusions on the assumption ofconstant diversification rates in the Ascomycota. The LTT plot(Fig. 3), suggest a slight declining trend in the number of accumu-lated ancestral lineages, but whether this is best explained byextinction events or by insufficient taxon sampling cannot bedetermined. The MCCR test supports these indications (resultsnot shown) though the number of replicates to achieve significantresults was insufficient.

It has been assumed that the marine and freshwater ascomy-cetes evolved from ancestors that occupied terrestrial habitats(Spatafora et al., 1998; Vijaykrishna et al., 2006). Around 530marine fungal species are known, 424 of which occur in variousorders of the Pezizomycotina (mostly members of Halosphaeri-ales, Spatafora et al., 1998; Jones et al., 2009). Additionally, 511freshwater Fungi are known in three Pezizomycotina classes:Leotiomycetes, Dothideomycetes and Sordariomycetes (Shearer,2001; Cai et al., 2003). Because marine Fungi occur in many dis-tinct Ascomycota lineages the question of a marine origin of thePezizomycotina is still disputed (e.g. Jones et al., 2009). However,

a marine origin is not well supported in recent literature, e.g. it isstill inconsistent with the ancestral reconstruction in Schoch et al.(2009a) and other recent studies (e.g. Sakayaroj, 2005; Schochet al., 2006).

The majority of the Pezizomycetes are terrestrial and live sap-rotrophically in soil. They typically build apothecia with opercu-late asci and it has been proposed that all other ascomata formsand spore release mechanisms have evolved from this type ofsporocarp with active spore release. According to our resultsthe Pezizomycetes diverged from other Pezizomycotina duringthe Ordovician. Since microbial mats including Fungi werealready present in the Proterozoic and fungal-like hyphae areknown from this period (Butterfield, 2005), a Neoproterozoicorigin of the Ascomycota and an Ordovician origin of thePezizomycotina is conceivable. Our results suggest that theLeotiomyceta (Pezizomycotina excluding Pezizomycetes andOrbiliomycetes) split from basal Pezizomycotina in the Silurianand subsequently diverged during the Devonian. This supportsa coevolutionary scenario of major land plant lineages and majorPezizomycotina lineages in the early Paleozoic. During the Devo-nian, main lineages of vascular plants (except, e.g. angiosperms)appeared, and the terrestrial vegetation changed from smallplants in the Early Devonian to the progymnosperm forests ofthe Late Devonian (Meyer-Berthaud et al., 2010). This entailedthe development of soils and distinct root systems, which mayhave onset the formation of new ecological niches of ascomyce-tes. Parasitic Pezizomycotina species may have evolved in aquaticor terrestrial Devonian habitats, for example together withvascular plants, algae or arthropods.

A recently discovered fossil lichen described by Honegger et al.(2013) appears to confirm the presence of lichen-forming fungisince the Early Devonian (415 Ma). Our results further indicatean initial diversification of Lecanoromycetes in the late Carbonifer-ous, which proceeded continuously, apparently unaffected by massextinction events and major global climatic changes. This scenariocorrelates with the global development of forest ecosystems sincethe Carboniferous. Beyond this it is difficult to relate the develop-ment of distinct Pezizomycotina classes to the evolution of otherorganisms (plants and/or animals) since almost all classes (withthe except for Orbiliomycetes and Laboulbeniomycetes) comprisea broad range of different life forms such as parasitic, lichen-form-ing and other symbiotic and saprophytic forms. According to ourresults Lichinomycetes and Coniocybomycetes appeared subse-quently in the Perminan, 274 Ma (197–379).

According to our results, the origins of many Lecanoromycetesgenera reach back to the Late Jurassic. Biatora, for instance,seems to represent an old lineage, which diverged from theBacidia-Phyllopsora clade some 148 million years ago (Fig. 2).Biatoria and Phyllopsora are closely related and share similarhabitat preferences, but are strictly allopatric, with Phyllopsorabeing restricted to tropical habitats and Biatora to temperateand cool regions of the Northern Hemisphere. Our resultslargely correlate with an assumed divergence of these generaabout 140–170 Ma due to expansion of the Tethys oceanseparating Laurasia from Gondwana (Printzen and Lumbsch,2000).

If we follow extant fungal lineages backwards in time we inev-itably arrive at the question of how old genera could be. Accordingto our data, most genera originated in the Mesozoic with some, likeHypocenomyces or Peltigera (Lecanoromycetes) extending back tothe Permian period. As the Ascomycota represents a vast group(�64,000 species), our data set represents only a fraction of allAscomycota species and does not allow precise interpretations ofthe appearances of particular genera. Additionally, we mustassume that the vast majority of Phanerozoic species is extinctand thus cannot be considered in molecular analyses.


4.5. Conclusions and outlook

Ambers and cherts have the potential to preserve delicate struc-tures with extraordinary quality. In this way fossil inclusions cansometimes be determined to genus level, allowing the preciseassignment of the fossils to recent phylogenies. Here we evaluateall available fossil ascomycetes representing the oldest reliable evi-dence of the respective extant lineages and discuss their suitabilityas minimum age constraints for molecular evolution studies. Weused fossil species from amber that are assignable to three Pezizo-mycotina classes in order to constrain a molecular clock for amulti-gene Ascomycota phylogeny. This is one of the first studiesto evaluate the impact of internal node constraints on models ofmolecular evolution for the Ascomycota. Comparison of analysesperformed using multiple-fossil calibration points vs. analysesusing only a sole minimum age constraint (Paleopyrenomycites)show that the use of fossil age constraints (even if relatively young)in terminal groups of three Pezizomycotina classes significantlyaffects the estimated divergence times of basal nodes and nodesof terminal groups of all Ascomycota lineages. Our estimated diver-gence times were exclusively based on internal age constrains(either one or five) but largely agree with estimates in recent stud-ies (Taylor and Berbee, 2006; Lücking et al., 2009; Gueidan et al.,2011; Prieto and Wedin, 2013).

According to our results (Fig. 2) the diversification of the Pezizo-mycotina started in the Ordovician, proceeded continuouslythroughout the Phanerozoic, and was largely unaffected by massextinction events. Lineages of extant ascomycetes typically possessa variety of different life forms in each lineage. Classes or even fam-ilies of ascomycetes may comprise both specialist and generalistspecies. We suggest that the diverse ecological strategies presentin ascomycete lineages allowed at least some members to survivemajor extinction events. Such a scenario has already been sug-gested to explain the phenomenon that many species but only arelatively low number of genera of vascular plants became extinctat the Cretaceous-Paleogene boundary (Upchurch et al., 2007). Fur-thermore, fungal spores are likely to survive unfavorable environ-mental conditions at times of extinction events (e.g. Sussman andDouthit, 1973).

Further phylogenetic studies, increased taxon sampling, and theintegration of a more comprehensive fossil record will generatemore reliable chronograms that will prove the hypothesis of a con-stant diversification of ascomycotan Fungi during the Phanerozoic.An ongoing screening of newly discovered ambers (Schmidt et al.,2006, 2010a) is accumulating determinable fossils of Fungi (includ-ing lichens) which are likely to further improve models of molecu-lar evolution for fungal phylogenies, especially for individualgroups of Fungi.

Acknowledgements

We would like to thank Jochen Heinrichs (Munich) and HaraldSchneider (London) for constructive discussions. Hans Kerp (Mün-ster), Raymond van der Ham (Leiden) and George O. Poinar, Jr.(Corvallis) kindly provided unpublished photomicrographs of thePaleopyrenomycites, Petropus, and Parmelia fossils, respectively.We thank Volker Arnold (Heide), Christel and Hans Werner Hoffe-ins (Hamburg), and Jörg Wunderlich (Hirschberg and der Bergs-traße) for providing amber specimens for this study. MarionKotrba and Dieter Doczkal (Munich) kindly arranged a loan ofthe Stigmatomyces specimen for imaging and Vincent Perrichot(Rénnes) provided the Metacapnodiaceae fossil from Charentesamber. Fieldwork and collection in southern New Caledonia werekindly permitted by the Direction de l’Environnement (ProvinceSud), permit n� 17778/DENV/SCB. This study is publication number

117 from the Courant Research Centre Geobiology that is fundedby the German Excellence Initiative.

Appendix A. Supplementary material

Supplementary data associated with this article can be found,in the online version, at http://dx.doi.org/10.1016/j.ympev.2014.04.024.

References

Amo de Paz, G., Cubas, P., Divakar, P.K., Lumbsch, H.T., Crespo, A., 2011. Origin anddiversification of major clades in Parmelioid lichens (Parmeliaceae,Ascomycota) during the Paleogene inferred by bayesian analysis. PLoS ONE 6,e28161.

Arx, J.A., 1981. The Genera of Fungi Sporulating in Pure Culture, third ed. J. Cramer,Vaduz.

Benton, M.J., Donoghue, P.C.J., Asher, R.J., 2009. Calibrating and constrainingmolecular clocks. In: Hedges, S.B., Kumar, S. (Eds.), Timetree of Life.University Press, Oxford, pp. 35–86.

Berbee, M.L., Taylor, J.W., 1993. Dating the evolutionary radiations of the true Fungi.Can. J. Bot. 71, 1114–1127.

Berbee, M.L., Taylor, J.W., 2007. Rhynie chert: a window into a lost world of complexplant–fungus interactions. New. Phyt. 174, 475–479.

Berbee, M.L., Taylor, J.W., 2010. Dating the molecular clock in Fungi – how close arewe? Fung. Biol. Rev. 24, 1–15.

Blackwell, M., 2011. The Fungi: 1, 2, 3 . . . 5.1 million species? Am. J. Bot. 98, 426–438.

Blumenstengel, H., Volland, L., Krutzsch, W., 1999. Zur Stratigraphie und Fazies desTertiärs im Bitterfelder Raum – Referenzprofile tertiärer Ablagerungen in denTagebauen Goitsche, Golpa-Nord und Göbern. Geologisches LandesamtSachsen-Anhalt, Halle.

Brazee, N.J., Marra, R.E., Wick, R.L., 2012. Genotypic diversity of Armillaria gallicafrom mixed oak forests in Massachusetts. Mycologia 104, 53–61.

Butterfield, N.J., 2005. Probable Proterozoic Fungi. Paleobiology 31, 165–182.Cai, L., Zhang, K.Q., Hyde, K.D., 2003. Freshwater ascomycetes. Fungal Divers 10,

275–324.Caspary, R., Klebs, R., 1907. Die Flora des Bernsteins und anderer fossiler Harze des

ostpreussischen Tertiärs. Berlin: Königlich Preußische GeologischeLandesanstalt, Neue Folge, Heft 4.

Crane, P.R., Friis, E.M., Pedersen, K.R., 1995. The origin and early diversification ofangiosperms. Nature 374, 27–33.

Darriba, D., Taboada, G.L., Doallo, R., Posada, D., 2012. JModelTest 2: more models,new heuristics and parallel computing. Nat. Methods 9, 772.

Doolittle, R.F., Feng, D.F., Tsang, S., Cho, G., Little, E., 1996. Determining divergencetimes of the major kingdoms of living organisms with a protein clock. Science271, 470–477.

Dörfelt, H., Schmidt, A.R., 2005. A fossil Aspergillus from Baltic amber. Mycol. Res.109, 956–960.

Dörfelt, H., Schmidt, A.R., 2007. A conifer seedling with two herbicolous Fungi fromthe Baltic amber forest. Bot. J. Lin. Soc. 155, 449–456.

Dornburg, A., Beaulie, J.M., Oliver, J.C., Near, T.J., 2011. Integrating fossilpreservation biases in the selection of calibrations for molecular divergencetime estimation. Syst. Biol. 60, 519–527.

Douzery, E.J.P., Snell, E.A., Bapteste, E., Delsuc, F., Philippe, H., 2004. The timing ofeukaryotic evolution: Does a relaxed molecular clock reconcile proteins andfossils? Proc. Natl. Acad. Sci. USA 101, 15386–15391.

Drummond, A.J., Rambaut, A., 2007. BEAST: Bayesian evolutionary analysis bysampling trees. BMC Evol. Biol. 7, 214.

Drummond, A.J., Ho, S.Y., Rambaut, A., 2006. Relaxed phylogenetics and dating withconfidence. PLoS. Biol. 4, e88.

Drummond, A.J., Suchard, M.A., Xie, Dong, Rambaut, A., 2012. Bayesianphylogenetics with BEAUti and the BEAST 1.7. Mol. Biol. Evol. 29, 1969–1973.

Dunlop, J.A., 2010. Bitterfeld amber. In: Penney, D. (Ed.), Biodiversity of Fossilsin Amber from the Major World Deposits. Siri Scientific Press, Manchester,pp. 57–68.

Ebersberger, I., de Matos Simoes, R., Kupczok, A., Gube, M., Kothe, E., Voigt, K., vonHaeseler, A., 2012. A consistent phylogenetic backbone for the Fungi. Mol. Biol.Evol. 29, 1319–1334.

Eriksson, O.E., 2005. Origin and evolution of Ascomycota—the protolicheneshypothesis. Svensk. Mykol. Tidskr. 26, 30–33.

Feldberg, K., Heinrichs, J., Schmidt, A.R., Vána, J., Schneider, H., 2013. Exploring theimpact of fossil constraints on the divergence time estimates of derivedliverworts. Plant. Syst. Evol. 299, 585–601.

Floudas, D. et al., 2012. The Paleozoic origin of enzymatic lignin decompositionreconstructed from 31 fungal genomes. Science 336, 1715–1719.

Forest, F., 2009. Calibrating the Tree of Life: fossils, molecules and evolutionarytimescales. Ann. Bot. 104, 789–794.

Gargas, A., Taylor, W.J., 1992. Polymerase chain reaction (PCR) primers foramplifying and sequencing nuclear 18S rDNA from lichenized Fungi.Mycologia 84, 589–592.



http://refhub.elsevier.com/S1055-7903(14)00152-3/h0005

































































Gouy, M., Guindon, S., Gascuel, O., 2010. Seaview version 4: a multiplatformgraphical user interface for sequence anignment and phylogenetic tree building.Mol. Biol. Evol. 27, 221–224.

Gueidan, C., Ruibal, C., de Hoog, S., Schneider, H., 2011. Rock-inhabiting Fungioriginated during periods of dry climate in the late Devonian and middleTriassic. Fungal Biol. 115, 987–996.

Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor andanalysis program for Windows 95/98/NT. Nucl. Acids. Symp. Ser 41, 95–98.

Hawksworth, D.L., 1991. The fungal dimension of biodiversity: magnitude,significance, and conservation. Mycol. Res. 95, 641–655.

Hawksworth, D.L., 2001. The magnitude of fungal diversity: 1.5 million speciesestimated revised. Mycol. Res. 105, 1422–1432.

Heckman, D.S., Geiser, D.M., Eidell, B.R., Stauffer, R.L., Kardos, N.L., Hedges Blair, S.,2001. Molecular evidence for the early colonization of land by Fungi and plants.Science 293, 1120–1133.

Hedman, M.H., 2010. Constraints on clade ages from fossil outgroups. Paleobiology36, 16–31.

Hibbett, D.S., Grimaldi, D., Donoghue, M.J., 1995. Cretaceous mushrooms in amber.Nature 377, 487.

Hibbett, D.S., Grimaldi, D., Donoghue, M.J., 1997. Fossil mushrooms from Mioceneand Cretaceous ambers and the evolution of Homobasidiomycetes. Am. J. Bot.84, 981–991.

Hibbett, D.S. et al., 2007. A higher-level phylogenetic classification of the Fungi.Mycol. Res. 111, 509–547.

Hofstetter, V., Miadlikowska, J., Kauff, F., Lutzoni, F., 2007. Phylogenetic comparisonof protein-coding versus ribosomal RNA-coding sequence data: a case study ofthe Lecanoromycetes (Ascomycota). Mol. Phyl. Evol. 44, 412–426.

Honegger, R., Edwards, D., Axe, L., 2013. The earliest records of internally stratifiedcyanobacterial and algal lichens from the Lower Devonian of the WelshBorderland. New Phyt. 196, 264–275.

Hustad, V.P., Miller, A.N., Dentinger, B.T.M., Cannon, P.F., 2013. Genericcircumscriptions in Geoglossomycetes. Persoonia 31, 101–111.

Inoue, J., Donoghue, P.C.J., Yang, Z., 2010. The impact of the representation of fossilcalibrations on Bayesian estimation of species divergence times. Syst. Biol. 59,74–89.

Iturralde-Vinent, M.A., 2001. Geology of the amber-bearing deposits of the GreaterAntilles. Carib. J. Sci. 37, 141–167.

Iturralde-Vinent, M.A., Mac Phee, R.D.E., 1996. Age and paleogeographical origin ofDominican amber. Science 273, 1850–1852.

Jagt, J.W.M., 1999. Late Cretaceous–early Palaeogene echinoderms and the K/Tboundary in the southeast Netherlands and northeast Belgium: Part 1.Introduction and stratigraphy. Scr. Geol. 116, 1–57.

James, T.Y. et al., 2006. Reconstructing the early evolution of Fungi using a six-genephylogeny. Nature 443, 818–822.

Jones, E.B.G., Sakayaroj, J., Suetrong, S., Somrithipol, S., Pang, K.L., 2009.Classification of marine Ascomycota, anamorphic taxa and Basidiomycota.Fungal Divers 35, 1–187.

Katoh, K., Toh, H., 2008. Recent developments in the MAFFT multiple sequencealignment program. Brief Bioinform. 92, 86–98.

Kirk, P.M., Cannon, P.F., Minter, D.W., Stalpers, J.A., 2008. Ainsworth and Bisby’sDictionary of the Fungi, tenth ed. CAB International, UK.

Kreader, C.A., 1996. Relief of amplification inhibition in PCR with bovine serumalbumin or T4 gene 32 protein. Appl. Environ. Microbiol. 62, 1102–1106.

Kumar, T.K.A., Healy, R., Spatafora, J.W., Blackwell, M., McLaughlin, D.J., 2012. Orbiliaultrastructure, character evolution and phylogeny of Pezizomycotina.Mycologia 104, 462–476.

Lara, E., Moreira, D., Lo9pez-García, P., 2010. The environmental clade LKM11 andRozella form the deepest branching clade of Fungi. Protist 161, 116–121.

Liu, Y.J., Hall, B.D., 2004. Body plan evolution of ascomycetes, as inferred from anRNA polymerase II phylogeny. Proc. Natl. Acad. Sci. USA 101, 4507–4512.

Liu, Y.J., Whelen, S., Hall, B.D., 1999. Phylogenetic relationships amongAscomycetes: evidence from an RNA Polymerse II Subunit. Mol. Biol. Evol. 16,1799–1808.

Liu, Y., Steenkamp, E.T., Brinkmann, H., Forget, L., Philippe, H., Lang, F., 2009.Phylogenomic analyses predict sistergroup relationship of nucleariids andFungi and paraphyly of zygomycetes with significant support. BMC Evol. Biol. 9,272.

Lücking, R., Huhndorf, S., Pfister, D.H., Plata, E.R., Lumbsch, T.H., 2009. Fungi evolvedright on track. Mycologia 10, 810–822.

Lukoschek, V., Keogh, J.S., Avise, J.C., 2012. Evaluating fossil calibrations for datingphylogenies in light of rates of molecular evolution: a comparison of threeapproaches. Syst. Biol. 61, 22–43.

Lumbsch, T.H., 2000. Phylogeny of filamentous ascomycetes. Naturwissenschaften87, 335–342.

Lumbsch, H.T., Hipp, A.L., Divakar, P.K., Blanco, O., Crespo, A., 2008. Acceleratedevolutionary rates in tropical and oceanic parmelioid lichens (Ascomycota).BMC Evol. Biol. (8), 257.

Lutzoni, F., Pagel, M., 1997. Accelerated evolution as a consequence of transitions tomutualism. Proc. Natl. Acad. Sci. USA 94, 11422–11427.

Lutzoni et al., 2004. Assembling the fungal tree of live: progress, classification, andevolution of subcellular traits. Am. J. Bot. 91, 1446–1480.

Mägdefrau, K., 1957. Flechten und Moose im Baltischen Bernstein. Ber. Dtsch. bot.Ges. 70, 433–435.

Magallon, S.A., 2010. Using fossils to break long branches in molecular dating: acomparison of relaxed clocks allied to the origin of angiosperms. Syst. Biol. 59,384–399.

Marshall, C.R., 2008. A simple method for bracketing absolute divergence times onmolecular phylogenies using multiple fossil calibration points. Am. Nat. 171,726–742.

Martínez-Delclo9s, X., Briggs, D.E.G., Penalver, E., 2004. Taphonomy of insects incarbonates and amber. Palaeogeogr. Palaeoclimatol. Palaeoecol. 203, 19–64.

Meyer-Berthaud, B., Soria, A., Decombeix, A.-L., 2010. Land plant cover in theDevonian: a reassessment of the evolution of the tree habit. In: Vecoli, M.,Clément, G., Meyer-Berthaud, B. (Eds.), The Terrestrialisation Process:Modelling Complex Interactions at the Biosphere/Geosphere Interface. Geol.Soc. London, Spec. Public. 339, pp. 59–70.

Miadlikowska, J. et al., 2006. New insights into classification and evolution of theLecanoromycetes (Pezizomycotina, Ascomycota) from phylogenetic analyses ofthree ribosomal RNA- and two protein-coding genes. Mycologia 98, 1088–1103.

Moncalvo, J.M., Lutzoni, F.M., Rehner, S., Johnson, J., Vilgalys, R., 2000. Phylogeneticrelationships of agaric Fungi based on nuclear large subunit ribosomal DNAsequences. Syst. Biol. 49, 278–305.

Morgenstern, I., Powlowski, J., Ishmael, N., Darmond, C., Marqueteau, S., Moisan,M.C., Quenneville, G., Tsang, A., 2012. A molecular phylogeny of thermophilicFungi. Fungal Biol. 116, 489–502.

Néraudeau, D., Perrichot, V., Dejax, J., Masure, E., Nel, A., Philippe, M., Moreau, P.,Guillocheau, F., Guyot, T., 2002. Un nouveau gisement à ambre insectifère et àvégétaux (Albien terminal probable): Archingeay (Charente-Maritime, France).Géobios 35, 233–240.

Ohm et al., 2012. Diverse lifestyles and strategies of plant pathogenesis encoded inthe genomes of eighteen Dothideomycetes Fungi. PLOS Pathog. 8, e1003037.

Padovan, A.C.B., Sanson, G.F.O., Brunstein, A., Briones, M.R.S., 2005. Fungi evolutionrevisited: application of the Penalized Likelihood Method to a Bayesian fungalphylogeny provides a new perspective on phylogenetic relationships anddivergence dates of Ascomycota groups. J. Mol. Evol. 60, 726–735.

Parham, J.F. et al., 2012. Best practices for justifying fossil calibrations. Syst. Biol. 61,346–359.

Perrichot, V., Néraudeau, D., Tafforeau, P., 2010. Charentese amber. In: Penney, D.(Ed.), Biodiversity of Fossils in Amber from the Major World Deposits. SiriScientific Press, Manchester, pp. 192–207.

Poinar Jr., G.O., Peterson, E.B., Platt, J.L., 2000. Fossil Parmelia in New World Amber.Lichenologist 32, 263–269.

Prieto, M., Wedin, M., 2013. Dating the Diversification of the major lineages ofAscomycota (Fungi). PLOSone 8, e65576.

Prieto, M., Baloch, E., Tehler, A., Wedin, M., 2013. Mazaedium evolution in theAscomycota (Fungi) and the classification of mazaediate groups of formerlyunclear relationship. Cladistics 29, 296–308.

Printzen, C., Lumbsch, T., 2000. Molecular evidence for the diversification ofextant lichens in the Late Cretaceous and Tertiary. Mol. Phylogenet. Evol. 17,379–387.

Pybus, O.G., Harvey, P.H., 2000. Testing macro-evolutionary models usingincomplete molecular phyogenies. Proc. Roy. Soc. Lond. B 267, 2267–2272.http://dx.doi.org/10.1098/rspb.2000.1278.

Pyron, R.A., 2010. A likelihood method for assessing molecular divergence timeestimates and the placement of fossil calibrations. Syst. Biol. 59, 185–195.

Rambaut, A., 2006–2009. FigTree. Tree figure drawing tool version 1.3.1, Institute ofEvolutionary Biology, University of Edinburgh. <http://tree.bio.ed.ac.uk/software/figtree/>.

Rambaut, A., Drummond, A.J., 2009. Tracer. MCMC Trace analysis tool versionv1.5.0. <http://tree.bio.ed.ac.uk/software/tracer/>.

Redecker, D., Kodner, R., Graham, L.E., 2000. Glomalean Fungi from the Ordovician.Science 289, 1920–1921.

Rehner, S.A., Samuels, G.J., 1994. Taxonomy and phylogeny of Gliocladium analyzedby large subunit rDNA sequences. Mycol. Res. 98, 625–634.

Rice, C.M., Ashcroft, W.A., Batten, D.J., Boyce, A.J., Caulfield, J.B.D., Fallick, A.E., Hole,M.J., Jones, E., Pearson, M.J., Rogers, G., Saxton, J.M., Stuart, F.M., Trewin, N.H.,Turner, G., 1995. A Devonian auriferous hot springs system, Rhynie, Scotland. J.Geol. Soc. Lond. 152, 229–250.

Richardson, J.B., 1967. Some British Lower Devonian spore assemblages and theirstratigraphic significance. Rev. Palaeobot. Palynol. 1, 111–129.

Rikkinen, J., 2003. Calicioid lichens from European Tertiary amber. Mycologia 95,1032–1036.

Rikkinen, J., Poinar, G., 2000. A new species of resinicolous Chaenothecopsis(Mycocaliciaceae, Ascomycota) from 20 million year old Bitterfeld amber,with remarks on the biology of resinicolous Fungi. Mycol. Res. 104, 7–15.

Rikkinen, J., Poinar, G., 2002. Fossilised Anzia (Lecanorales, lichen-formingAscomycota) from European Tertiary amber. Mycol. Res. 106, 984–990.

Rikkinen, J., Poinar, G., 2008. A new species of Phyllopsora (Lecanorales, lichen-forming Ascomycota) from Dominican amber, with remarks on the fossilhistory of lichens. J. Exp. Bot. 59 (5), 1007–1011.

Rikkinen, J., Dörfelt, H., Schmidt, A.R., Wunderlich, J., 2003. Sooty moulds fromEuropean Tertiary amber, with notes on the systematic position of Rosaria(Cyanobacteria). Mycol. Res. 107, 251–256.

Ronquist, F., Huelsenbeck, J.P., 2003. Bayesian phylogenetic inference under mixedmodels. Bioinformatics 19, 1572–1574.

Rossi, W., Kotrba, M., Triebel, D., 2005. A new species of Stigmatomyces from Balticamber, the first fossil record of Laboulbeniomycetes. Mycol. Res. 109, 271–274.

Rutschmann, F., Eriksson, T., Salim, K.A., Conti, E., 2007. Assessing calibrationuncertainty in molecular dating: the assignments of fossils to alternativecalibration points. Syst. Biol. 56, 591–608.

Sakayaroj, J., 2005. Phylogenetic relationships of marine Ascomycota. Ph.D. Thesis,Prince of Songkla University, Thailand.





































































































































http://dx.doi.org/10.1098/rspb.2000.1278





http://tree.bio.ed.ac.uk/software/tracer/
































Sanderson, M.J., 2003a. Molecular data from 27 proteins do not support aPrecambrian origin of land plants. Am. J. Bot. 90, 954–956.

Sanderson, M.J., 2003b. R8s: inferring absolute rates of molecular evolution anddivergence times in the absence of a molecular clock. Bioinformatics 19, 301–302.

Sauquet, H. et al., 2012. Testing the impact of calibration on molecular divergencetimes using a fossil-rich group: the case of Nothofagus (Fagales). Syst. Biol. 61,289–313.

Schmidt, A.R., Ragazzi, E., Coppellotti, O., Roghi, G., 2006. A microworld in Triassicamber. Nature 444, 835.

Schmidt et al., 2010a. Cretaceous African life captured in amber. Proc. Natl. Acad.Sci. USA 107, 7329–7334.

Schmidt, A.R., Dörfelt, H., Struwe, S., Perrichot, V., 2010b. Evidence for Fungivory inCretaceous amber forests from Gondwana and Laurasia. Palaeontographica 283,157–173.

Schmidt, A.R., Beimforde, C., Seyfullah, L.J., Wege, S.E., Dörfelt, H., Girard, V.,Grabenhorst, H., Gube, M., Heinrichs, J., Nel, A., Nel, P., Perrichot, V., Reitner, J.,Rikkinen, J., 2014. Amber fossils of sooty moulds. Rev. Palaeobot. Palynol. 200,53–64.

Schoch, C.L., Sung, G.-H., Volkmann-Kohlmeyer, B., Kohlmeyer, J., Spatafora, J.W.,2006. Marine fungal lineages in the Hypocreomycetidae. Mycol. Res. 110, 257–263.

Schoch et al., 2009a. The Ascomycota Tree of Life: a phylum-wide phylogenyclarifies the origin and evolution of fundamental reproductive and ecologicaltraits. Syst. Biol. 58, 224–239.

Schoch, C.L., Wang, Z., Townsend, J.P., Spatafora, J.W., 2009b. Geoglossomycetes cl.nov., Geoglossales ord. nov. and taxa above class rank in the Ascomycota Tree ofLife. Persoonia 22, 129–138.

Schwarz, G., 1978. Estimating the dimension of a model. Ann. Stat. 6, 461–464.Shearer, C.A., 2001. The distribution of freshwater filamentous Ascomycetes. In:

Misra, J.K., Horn, B.W. (Eds.), Trichomycetes and Other Fungal Groups: RobertW. Lichtwardt Commemoration Volume. Science Publishers Inc., Enfield, NewHampshire, pp. 225–292.

Simon, L., Bousquet, J., Levesque, R.C., Lalonde, M., 1993. Origin and diversificationof endomycorrhizal Fungi and coincidence with vascular land plants. Nature363, 67–69.

Spatafora, J.W., Volkmann-Kohlmeyer, B., Kohlmeyer, J., 1998. Independentterrestrial origins of the Halosphaeriales (Marine Ascomycota). Am. J. Bot. 85,1569–1580.

Spatafora, J.W. et al., 2006. A five-gene phylogenetic analysis of the Pezizomycotina.Mycologia 98, 1020–1030.

Stadler, T., 2009. On incomplete sampling under birth-death models andconnections to the sampling-based coalescent. J. Theor. Biol. 261, 58–66.

Stamatakis, A., Hoover, J., Rougemint, J., 2008. A rapid Bootstrap algorithm for theRAxML Web Servers. Syst. Biol. 75, 558–771.

Standke, G., 2008. Bitterfelder Bernstein gleich Baltischer Bernstein? – Einegeologische Raum-Zeit-Betrachtung und genetische Schlußfolgerungen.Exkurs.f. und Veröfftl. DGG 236, 11–33.

Stankiewicz, B.A., Poinar, H.N., Briggs, D.E.G., Evershed, R.P., Poinar Jr., G.O., 1998.Chemical preservation of plants and insects in natural resins. Philos. Trans. Roy.Soc. B 265, 641–647.

Steenkamp, E.T., Wright, J., Baldauf, S.L., 2006. The protistan origins of animals andFungi. Mol. Biol. Evol. 23, 93–106.

Sussman, A.S., Douthit, H.A., 1973. Dormancy in microbial spores. Ann. Rev. Plant.Phylsiol. 24, 311–352.

Taylor, J.W., Berbee, M.L., 2006. Dating divergences in the Fungal Tree of Life: reviewand new analyses. Mycologia 98, 838–849.

Taylor, T.N., Remy, W., Hass, H., 1994. Allomyces in the Devonian. Nature 367, 601.Taylor, T.N., Hass, H., Kerp, H., 1999. The oldest fossil ascomycete. Nature 399,

648.Taylor, T.N., Hass, H., Kerp, H., Krings, M., Hanlin, R.T., 2005. Perithecial ascomycetes

from the 400 million year old Rhynie chert: an example of ancestralpolymorphism. Mycologia 97, 269–285.

Thell, A., Crespo, A., Divakar, P.K., Kärnefelt, I., Leavitt, S.D., Lumbsch, T.H., Seaward,M.R.D., 2012. A review of the lichen family Parmeliaceae – history, phylogenyand current taxonomy. Nord. J. Bot. 30, 641–664.

Tibell, L., 1984. A reappraisal of the taxonomy of caliciales. Beiheft zur NovaHedwigia 79, 597–713.

Tibell, L., 2001. Photobiont association and molecular phylogeny of the lichen genusChaenotheca. Bryologist 104, 191–198.

Tibell, L., Koffman, A., 2002. Chaenotheca nitidula, a new species of calicioid lichenfrom Northeastern North America. Bryologist 105, 353–357.

Tuovila, H., Schmidt, A.R., Beimforde, C., Dörfelt, H., Grabenhorst, H., Rikkinen, J.,2013. Stuck in time – a new Chaenothecopsis species with proliferatingascomata from Cunninghamia resin and its fossil ancestors in Europeanamber. Fungal Divers. 58, 199–213.

Upchurch, G.R., Lomax, B.H., Beerling, D.J., 2007. Paleobotanical evidence forclimatic change across the Cretaceous-Tertiary Boundary, North America:twenty years after Wolfe and Upchurch. Cour. Forsch.-Inst. Senckenberg 258,57–74.

Van der Ham, R.W.J.M., Dortangs, R.W., 2005. Structurally preserved ascomycetousFungi from the Maastrichtian type area (NE Belgium). Rev. Palaeobot. Palynol.136, 48–62.

Van der Ham, R.W.J.M., Van Konijnenburg-van Cittert, J.H.A., 2003. Rare conifersfrom the type area of the Maastrichtian (Upper Cretaceous SoutheastNetherlands). Scr. Geol. 126, 111–119.

Vijaykrishna, D., Jeewon, R., Hyde, K.D., 2006. Molecular taxonomy, origins andevolution of freshwater ascomycetes. Fungal Divers 23, 351–390.

Vilgalys, R., Hester, M., 1990. Rapid genetic identification and mapping ofenzymatically amplified ribosomal DNA from several Cryptococcus species. J.Bacteriol. 172, 4238–4246.

Wilkinson, R.D., Steiper, M.E., Soligo, C., Martin, R.D., Yang, Z.H., Tavaré, S., 2011.Dating primate divergences through an integrated analysis of palaeontologicaland molecular data. Syst. Biol. 60, 16–31.

Winton, L.M., Stone, J.K., Hansen, E.M., 2007. The systematic position ofPhaeocryptopus gaeumannii. Mycologia 99, 240–252.

White, T.J., Bruns, T., Lee, S., Taylor, J., 1990. Amplification and direct sequencing offungal ribosomal RNA genes for phylogenetics. In: Innis, M.A., Gelfand, D.H.,Sninsky, J.J., White, T.J. (Eds.), PCR Protocols. Academic Press, San Diego, pp.315–322.

Woolfit, M., Bromham, L., 2003. Increased rates of sequence evolution inendosymbiotic bacteria and Fungi with small effective population sizes. Mol.Biol. Evol. 20, 1545–1555.

Zhang, Y., Crous, P., Schoch, C., Bahkali, A., Guo, L., Hyde, K., 2011. A molecular,morphological and ecological re-appraisal of Venturiales, a new order ofDothideomycetes. Fungal Div. 51, 249–277.











































































































Estimating the Phanerozoic history of the Ascomycota ...

Documents

Transcript of Estimating the Phanerozoic history of the Ascomycota ...